Skip to contents

Plot Cell Proportions

plotCellProportions.Rd

Takes a SpatialMap or Region object, extracts cellMetadata, and plots the proportions of 'var1' cells across 'var2'. Var1 will be treated as separate groups, and Var2 will be represented as fill colors. Any output can be modified as a ggplot object. Examples shown below.

Usage

plotCellProportions(
  object,
  var1,
  var2,
  type = "bar",
  pos = "auto",
  filter.eval = NULL,
  var_sample = "Region",
  select_var1 = NULL,
  select_var2 = NULL,
  sample_props = TRUE
)

Arguments

object

SpatialMap or Region object

var1

Variable name from cellMetadata or projectMetadata. Defines groups on the x axis (aes(x = var1) to paraphrase ggplot2 syntax).

var2

Variable name from cellMetadata or projectMetadata. Defines groups that will be indicated by different fill colors (aes(fill = var2) to paraphrase ggplot2 syntax).

type

(optional) Specifies type of graph to create. Graph type can be "bar", "box", "violin", or "strip_chart" (default="bar").

pos

(optional) position argument passed to geom_* function. For example, if type=="bar" then pos is passed to geom_bar as geom_bar(..., position = pos).

filter.eval

A quoted character string that will be parsed and evaluated by rlang::parse_expr within a dplyr filter

var_sample

(optional) variable used to demarcate samples for strip charts, violin charts, or box charts(default="Region")

select_var1

Used to select which items from var1 should be included in the graph. (default = NULL, meaning no items excluded from final chart).

select_var2

Same as "select_var1", except for var2.

sample_props

(optional) Whether to plot cell proportions within var_sample (default = TRUE). If TRUE and type=="bar", the mean cell proportions will be plotted. If FALSE, cell counts will be plotted instead.

Value

A ggplot object

Details

See vignette("AnalysisGuide7_Cohort_comparison") for an example workflow that uses this function.

Examples

library(ggplot2)
sm <- load_sm_data("skin")
sm <- mergeProjectMetadata(sm)
plotCellProportions(sm,
                    var1 = "cell_type",
                    var2 = "dermatitis_status")


# Add your own ggplot parameters and graph titles
 plotCellProportions(sm,
                     var1 = "cell_type",
                     var2 = "dermatitis_status",
                     type = "bar") +
  labs(title = "Cell type proportions colored by disease status in bar chart",
       x = "Cell Type",
       y = "Count")


# Plot multiple types of graphs by using `type` parameter
# `type` can be set to: "bar", "box", "violin", or "strip_chart"
plotCellProportions(sm,
                    var1 = "cell_type",
                    var2 = "dermatitis_status",
                    type = "strip_chart",
                    select_var1 = c("B cells (CD20+)", "Macs (CD68+)")) +
   labs(title = "Cell type proportions colored by disease status in strip chart",
        x = "Cell Type",
        y = "Count")
Scroll to top